From: "Alex Bennée" <alex.bennee@linaro.org>
To: qemu-devel@nongnu.org
Cc: f4bug@amsat.org, "Alex Bennée" <alex.bennee@linaro.org>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Peter Xu" <peterx@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Marcel Apfelbaum" <marcel.apfelbaum@gmail.com>
Subject: [PATCH v5 17/20] hw/intc: properly model IOAPIC MSI messages
Date: Fri, 11 Nov 2022 18:25:32 +0000 [thread overview]
Message-ID: <20221111182535.64844-18-alex.bennee@linaro.org> (raw)
In-Reply-To: <20221111182535.64844-1-alex.bennee@linaro.org>
On the real HW the IOAPIC is wired directly to the APIC and doesn't
really generate memory accesses on the main bus of the system. To
model this we can use the MTRT_MACHINE requester type and set the id
as a magic number to represent the IOAPIC as the source.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
---
include/hw/i386/ioapic_internal.h | 2 ++
hw/intc/ioapic.c | 35 ++++++++++++++++++++++++-------
2 files changed, 30 insertions(+), 7 deletions(-)
diff --git a/include/hw/i386/ioapic_internal.h b/include/hw/i386/ioapic_internal.h
index 9880443cc7..a8c7a1418a 100644
--- a/include/hw/i386/ioapic_internal.h
+++ b/include/hw/i386/ioapic_internal.h
@@ -82,6 +82,8 @@
#define IOAPIC_VER_ENTRIES_SHIFT 16
+/* Magic number to identify IOAPIC memory transactions */
+#define MEMTX_IOAPIC 0xA71C
#define TYPE_IOAPIC_COMMON "ioapic-common"
OBJECT_DECLARE_TYPE(IOAPICCommonState, IOAPICCommonClass, IOAPIC_COMMON)
diff --git a/hw/intc/ioapic.c b/hw/intc/ioapic.c
index 264262959d..8a5418002b 100644
--- a/hw/intc/ioapic.c
+++ b/hw/intc/ioapic.c
@@ -21,6 +21,7 @@
*/
#include "qemu/osdep.h"
+#include "qemu/log.h"
#include "qapi/error.h"
#include "monitor/monitor.h"
#include "hw/i386/apic.h"
@@ -88,9 +89,33 @@ static void ioapic_entry_parse(uint64_t entry, struct ioapic_entry_info *info)
(info->delivery_mode << MSI_DATA_DELIVERY_MODE_SHIFT);
}
-static void ioapic_service(IOAPICCommonState *s)
+/*
+ * No matter whether IR is enabled, we translate the IOAPIC message
+ * into a MSI one, and its address space will decide whether we need a
+ * translation.
+ *
+ * As the IOPIC is directly wired to the APIC writes to it are not the
+ * same as writes coming from the main bus of the machine. To model
+ * this we set its source as machine specific with the MEMTX_IOPIC
+ * id.
+ */
+static void send_ioapic_msi(struct ioapic_entry_info info)
{
AddressSpace *ioapic_as = X86_MACHINE(qdev_get_machine())->ioapic_as;
+ MemTxAttrs attrs = MEMTXATTRS_MACHINE(MEMTX_IOAPIC);
+ MemTxResult res;
+
+ address_space_stl_le(ioapic_as, info.addr, info.data,
+ attrs, &res);
+ if (res != MEMTX_OK) {
+ qemu_log_mask(LOG_GUEST_ERROR,
+ "%s: couldn't write to %"PRIx32"\n", __func__, info.addr);
+ }
+}
+
+
+static void ioapic_service(IOAPICCommonState *s)
+{
struct ioapic_entry_info info;
uint8_t i;
uint32_t mask;
@@ -130,12 +155,8 @@ static void ioapic_service(IOAPICCommonState *s)
continue;
}
#endif
-
- /* No matter whether IR is enabled, we translate
- * the IOAPIC message into a MSI one, and its
- * address space will decide whether we need a
- * translation. */
- stl_le_phys(ioapic_as, info.addr, info.data);
+ /* If not handled by KVM we now send it ourselves */
+ send_ioapic_msi(info);
}
}
}
--
2.34.1
next prev parent reply other threads:[~2022-11-11 18:28 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-11 18:25 [PATCH for 8.0 v5 00/20] use MemTxAttrs to avoid current_cpu in hw/ Alex Bennée
2022-11-11 18:25 ` [PATCH v5 01/20] hw: encode accessing CPU index in MemTxAttrs Alex Bennée
2022-11-12 4:18 ` Richard Henderson
2022-11-21 18:32 ` Peter Maydell
2022-11-11 18:25 ` [PATCH v5 02/20] target/arm: ensure TCG IO accesses set appropriate MemTxAttrs Alex Bennée
2022-11-12 5:17 ` Richard Henderson
2022-11-12 5:26 ` Richard Henderson
2022-11-11 18:25 ` [PATCH v5 03/20] target/arm: ensure HVF traps " Alex Bennée
2022-11-11 18:25 ` [PATCH v5 04/20] target/arm: ensure KVM " Alex Bennée
2022-11-12 5:29 ` Richard Henderson
2022-11-11 18:25 ` [PATCH v5 05/20] target/arm: ensure m-profile helpers " Alex Bennée
2022-11-12 5:26 ` Richard Henderson
2022-11-11 18:25 ` [PATCH v5 06/20] qtest: make read/write operation appear to be from CPU Alex Bennée
2022-11-11 18:25 ` [PATCH v5 07/20] hw/intc/gic: use MxTxAttrs to divine accessing CPU Alex Bennée
2022-11-11 18:25 ` [PATCH v5 08/20] hw/timer: convert mptimer access to attrs to derive cpu index Alex Bennée
2022-11-11 18:25 ` [PATCH v5 09/20] hw/arm: remove current_cpu hack from pxa2xx access Alex Bennée
2022-11-12 5:36 ` Richard Henderson
2022-11-13 19:43 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 10/20] target/microblaze: initialise MemTxAttrs for CPU access Alex Bennée
2022-11-11 19:41 ` Edgar E. Iglesias
2022-11-12 5:37 ` Richard Henderson
2022-11-13 19:44 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 11/20] target/sparc: " Alex Bennée
2022-11-12 1:02 ` Mark Cave-Ayland
2022-11-12 5:38 ` Richard Henderson
2022-11-13 19:45 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 12/20] target/riscv: " Alex Bennée
2022-11-11 18:25 ` [PATCH v5 13/20] target/i386: add explicit initialisation for MexTxAttrs Alex Bennée
2022-11-12 5:49 ` Richard Henderson
2022-11-11 18:25 ` [PATCH v5 14/20] hw/audio: explicitly set .requester_type for intel-hda Alex Bennée
2022-11-12 5:50 ` Richard Henderson
2022-11-13 19:50 ` Philippe Mathieu-Daudé
2022-11-21 18:39 ` Peter Maydell
2022-11-21 22:14 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 15/20] hw/i386: update vapic_write to use MemTxAttrs Alex Bennée
2022-11-12 5:51 ` Richard Henderson
2022-11-13 19:52 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 16/20] include: add MEMTXATTRS_MACHINE helper Alex Bennée
2022-11-12 5:52 ` Richard Henderson
2022-11-11 18:25 ` Alex Bennée [this message]
2022-11-12 5:57 ` [PATCH v5 17/20] hw/intc: properly model IOAPIC MSI messages Richard Henderson
2022-11-11 18:25 ` [PATCH v5 18/20] hw/i386: convert apic access to use MemTxAttrs Alex Bennée
2022-11-12 6:02 ` Richard Henderson
2022-11-21 18:43 ` Peter Maydell
2022-11-11 18:25 ` [PATCH v5 19/20] hw/isa: derive CPUState from MemTxAttrs in apm_ioport_writeb Alex Bennée
2022-11-12 6:04 ` Richard Henderson
2022-11-13 20:04 ` Philippe Mathieu-Daudé
2022-11-11 18:25 ` [PATCH v5 20/20] include/hw: add commentary to current_cpu export Alex Bennée
2022-11-12 6:05 ` Richard Henderson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20221111182535.64844-18-alex.bennee@linaro.org \
--to=alex.bennee@linaro.org \
--cc=f4bug@amsat.org \
--cc=marcel.apfelbaum@gmail.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peterx@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).